I've never been in the need of adding/updating/replacing license headers into every source code file in a big project, but now I am. And I always thought there was some popular tool for doing it, because many historical source tarballs have license headers in all their files together with the typical version header applied by the (then popular) RCS/SCCS version control systems.

I mean, just look at big sources published in the 90s by SGI, HP, Berkeley, etc, and it's there. I always thought that was automated, and not only for adding the license notice, but for updating/replacing/removing it.

But I searched and searched, and found no such tool. There are some tools for doing this at GitHub, but they are pretty recent... nothing with a >20-year tradition.

Did all the companies from the commercial UNIX era use their own unpublished in-house shell scripts for doing this? Wasn't there any popular tool for this in those years?

  • 1
    Likely those were in fact unpublished scripts, which usually simply replaced existing notices and dates with "updated" ones (I've actually written tools which do it right -- unpublished -- but see this for comments on the usual case). – Thomas Dickey Jul 26 at 21:57
  • @ThomasDickey I didn't know about your copyrite tool, and I really like it. The PDF manpage at the website, however, mentions it can only process 7bit chars, while some of my sources are UTF8-encoded. Do you think I could easily move the 1-bit mask out of the text data, so that UTF8-encoded files can be processed, or would it imply to perform great changes into the code? – cesss Jul 27 at 10:44
  • I think it would take a lot of work (I haven't tried). – Thomas Dickey Jul 27 at 11:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.